Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Couple typos and is_m4 function #498

Merged
merged 5 commits into from
Dec 17, 2024
Merged

Couple typos and is_m4 function #498

merged 5 commits into from
Dec 17, 2024

Conversation

christiangnrd
Copy link
Contributor

No description provided.

lib/mtl/device.jl Outdated Show resolved Hide resolved
@christiangnrd christiangnrd force-pushed the cleanup branch 2 times, most recently from 08d1ccb to f4d13d7 Compare December 16, 2024 17:01
@christiangnrd
Copy link
Contributor Author

@maleadt Do you know why the github actions benchmark test still triggers?

@maleadt
Copy link
Member

maleadt commented Dec 16, 2024

No. Did you verify it worked before? Maybe I messed something up in 6ecb909

@christiangnrd christiangnrd force-pushed the cleanup branch 4 times, most recently from f31fa7e to 175a9f6 Compare December 16, 2024 18:52
@christiangnrd
Copy link
Contributor Author

@maleadt Sorry for the constant notification spam. I'm done for now. I don't think you broke anything, head_commit doesn't seem to exist for pull_request events anymore. It's coming back green and it works for pushes to master (where it actually matters) so I won't spend any more time figuring it out on the PR.

@christiangnrd
Copy link
Contributor Author

Not sure what the Permission Denied errors from mac mini 5.0 are about.

@maleadt
Copy link
Member

maleadt commented Dec 17, 2024

Not sure what the Permission Denied errors from mac mini 5.0 are about.

The compilecache files on there have really weird permissions:

julia@macmini-aarch64-5 ~ % ls -la /Users/julia/.julia/scratchspaces/a66863c6-20e8-4ff4-8a62-49f30b1f605e/agent-cache/default-macmini-aarch64-5.0/depots/5cd495a2-4a16-4674-ae02-c839447744bb/compiled/v1.11/Metal/ACDsk_tBsmV.ji
--w-------  1 julia  staff  21152 Dec 16 13:28 /Users/julia/.julia/scratchspaces/a66863c6-20e8-4ff4-8a62-49f30b1f605e/agent-cache/default-macmini-aarch64-5.0/depots/5cd495a2-4a16-4674-ae02-c839447744bb/compiled/v1.11/Metal/ACDsk_tBsmV.ji

It's the only file with such permissions in there:

julia@macmini-aarch64-5 ~ % find /Users/julia/.julia/scratchspaces/*/agent-cache/*/depots/*/compiled -perm 200
/Users/julia/.julia/scratchspaces/a66863c6-20e8-4ff4-8a62-49f30b1f605e/agent-cache/default-macmini-aarch64-5.0/depots/5cd495a2-4a16-4674-ae02-c839447744bb/compiled/v1.11/Metal/ACDsk_tBsmV.ji

I'll clear it out, and I've filed JuliaLang/Pkg.jl#4118.

Copy link
Contributor

@github-actions github-actions bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Metal Benchmarks

Benchmark suite Current: bcf2d39 Previous: 5056e33 Ratio
private array/construct 26851.14285714286 ns 25920.083333333336 ns 1.04
private array/broadcast 467959 ns 464792 ns 1.01
private array/random/randn/Float32 835354 ns 811812 ns 1.03
private array/random/randn!/Float32 661375 ns 673333 ns 0.98
private array/random/rand!/Int64 566375 ns 553250 ns 1.02
private array/random/rand!/Float32 603937 ns 601000 ns 1.00
private array/random/rand/Int64 724666 ns 763896 ns 0.95
private array/random/rand/Float32 613875 ns 616375 ns 1.00
private array/copyto!/gpu_to_gpu 686458 ns 660854.5 ns 1.04
private array/copyto!/cpu_to_gpu 668041.5 ns 619854 ns 1.08
private array/copyto!/gpu_to_cpu 834542 ns 834687.5 ns 1.00
private array/accumulate/1d 1313584 ns 1328542 ns 0.99
private array/accumulate/2d 1389083 ns 1387334 ns 1.00
private array/iteration/findall/int 2081708 ns 2070500 ns 1.01
private array/iteration/findall/bool 1821083 ns 1824916 ns 1.00
private array/iteration/findfirst/int 1708041.5 ns 1682146 ns 1.02
private array/iteration/findfirst/bool 1661583 ns 1637959 ns 1.01
private array/iteration/scalar 3674250 ns 3891833 ns 0.94
private array/iteration/logical 3185292 ns 3177708.5 ns 1.00
private array/iteration/findmin/1d 1738375 ns 1740229 ns 1.00
private array/iteration/findmin/2d 1333959 ns 1343375 ns 0.99
private array/reductions/reduce/1d 1044500 ns 1035708 ns 1.01
private array/reductions/reduce/2d 663750 ns 651667 ns 1.02
private array/reductions/mapreduce/1d 1040625 ns 1037167 ns 1.00
private array/reductions/mapreduce/2d 660916 ns 657917 ns 1.00
private array/permutedims/4d 2554791.5 ns 2537625 ns 1.01
private array/permutedims/2d 1026771 ns 1022708 ns 1.00
private array/permutedims/3d 1587833 ns 1577291.5 ns 1.01
private array/copy 597792 ns 621750 ns 0.96
latency/precompile 5252928875 ns 5243893542 ns 1.00
latency/ttfp 6654524646 ns 6538101604 ns 1.02
latency/import 1162990312.5 ns 1165440583 ns 1.00
integration/metaldevrt 709416.5 ns 705833 ns 1.01
integration/byval/slices=1 1571333 ns 1588833.5 ns 0.99
integration/byval/slices=3 11345104 ns 10079959 ns 1.13
integration/byval/reference 1626041 ns 1568145.5 ns 1.04
integration/byval/slices=2 2642084 ns 2643625 ns 1.00
kernel/indexing 490583 ns 444959 ns 1.10
kernel/indexing_checked 475271 ns 446541 ns 1.06
kernel/launch 10090.166666666668 ns 10666.666666666666 ns 0.95
metal/synchronization/stream 14583 ns 15167 ns 0.96
metal/synchronization/context 15209 ns 15833 ns 0.96
shared array/construct 25375 ns 25503.5 ns 0.99
shared array/broadcast 481792 ns 475125 ns 1.01
shared array/random/randn/Float32 765187.5 ns 752770.5 ns 1.02
shared array/random/randn!/Float32 664000 ns 655458 ns 1.01
shared array/random/rand!/Int64 563750 ns 560916 ns 1.01
shared array/random/rand!/Float32 602958 ns 598459 ns 1.01
shared array/random/rand/Int64 778292 ns 774208 ns 1.01
shared array/random/rand/Float32 614020.5 ns 629583.5 ns 0.98
shared array/copyto!/gpu_to_gpu 88750 ns 85500 ns 1.04
shared array/copyto!/cpu_to_gpu 88416 ns 91792 ns 0.96
shared array/copyto!/gpu_to_cpu 81917 ns 77917 ns 1.05
shared array/accumulate/1d 1340166.5 ns 1347021 ns 0.99
shared array/accumulate/2d 1393791 ns 1383874.5 ns 1.01
shared array/iteration/findall/int 1774000 ns 1783167 ns 0.99
shared array/iteration/findall/bool 1577271 ns 1585458 ns 0.99
shared array/iteration/findfirst/int 1398166.5 ns 1392020.5 ns 1.00
shared array/iteration/findfirst/bool 1348542 ns 1354916.5 ns 1.00
shared array/iteration/scalar 157125 ns 162042 ns 0.97
shared array/iteration/logical 2976833 ns 2968958 ns 1.00
shared array/iteration/findmin/1d 1455187.5 ns 1461729.5 ns 1.00
shared array/iteration/findmin/2d 1361313 ns 1364792 ns 1.00
shared array/reductions/reduce/1d 740125 ns 730458 ns 1.01
shared array/reductions/reduce/2d 654333 ns 656958 ns 1.00
shared array/reductions/mapreduce/1d 744750 ns 746250 ns 1.00
shared array/reductions/mapreduce/2d 668166 ns 660562.5 ns 1.01
shared array/permutedims/4d 2546875 ns 2528541.5 ns 1.01
shared array/permutedims/2d 1026062.5 ns 1025917 ns 1.00
shared array/permutedims/3d 1594500 ns 1580896 ns 1.01
shared array/copy 237334 ns 242437.5 ns 0.98

This comment was automatically generated by workflow using github-action-benchmark.

@maleadt maleadt merged commit d37e9dd into main Dec 17, 2024
2 checks passed
@maleadt maleadt deleted the cleanup branch December 17, 2024 12:19
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants